The often-stated policy goals of the Federal Reserve’s policy making group (the FOMC) are “to achieve maximum employment and inflation at the rate of 2 percent over the longer run”. These goals stem from the Federal Reserve Reform Act of 1977, legislation passed by Congress almost 50 years ago when inflation was 6.5 percent and climbing, the unemployment rate was 6.8 percent, and the 10-year Treasury yield was 7.8 percent. The Act directs the Federal Reserve to conduct U.S. monetary policy “so as to promote the goals of maximum employment, stable prices, and moderate long-term interest rates.” The message from Congress to the Federal Reserve was clear: We want all these things — inflation, unemployment and long-term interest rates — to be lower … a lot lower.
But the Act didn’t quantify how much lower. The 2 percent inflation goal was a creation of the Federal Reserve, not Congress. It was formally adopted at the start of 2012 when Ben Bernanke was Chair of FOMC and the country was slowly emerging from the financial crisis of 2007-2009. Inflation was then 2.6 percent and falling, unemployment was 8.3 percent, and the 10-year Treasury yield was 2.0 percent. In that economic setting, a “2 percent” inflation target probably didn’t seem too ambitious. But it would be wrong to say the 2 percent target was based on a body of research that demonstrated that it was either optimal or even feasible in the “long run”. It was not.
What we do know is 2 percent is not the “normal” rate of inflation in America. Since the end of the 1940s, the 12-month inflation rate has averaged 3.5 percent (measured by the CPI) or 3.1 percent (using the PCE deflator) – both well above 2 percent (green line, chart below). To get an inflation rate that averages 2 percent “over the longer run” implies, statistically, that measured inflation is less than 2 percent for many of the years that make up “the longer run”. Very low (below 2 percent) inflation years in the U.S. have tended to cluster together, and there have only been two such “clusters” in the past 75 years. One low inflation “cluster” occurred between the end of the Korean war in 1953 and the mid-1960s, and the other happened following the financial crisis when the Fed’s formal 2 percent target was established.
Clustering is a common feature in price data. This makes sense, since the prices of goods and services are set by people, and some peoples’ pricing decisions strongly influence the pricing behavior of others, particularly the pricing decisions made by people that manage companies in industries that are important in American society. But pricing decisions by business leaders don’t occur in a vacuum. Government policies, at both the Federal and state & local levels, play a central role in determining whether America is in a high-inflation or low-inflation environment, as does the relative power of labor. What is going on in the world outside America also plays a role. Wars are often associated with high inflation, since they tend to be associated with disruptions in cross-border trade flows and commodity production, as well as increased defense spending by nation states.
Supply, not demand, is the problem
When we look at American society today — and the country’s economic and political place in the world – there appear to more differences than similarities compared with the environments that produced the clusters of low-inflation years in the past, differences that we believe will generate above-average inflation in coming years. In the pages below, we’ll describe the conditions that created the pricing behaviors in the two periods of sustained below-average inflation and contrast them with the situation we find ourselves in today. What particularly stands out now in America are the existing shortages of housing, of labor, and of electricity production to meet current and future demands in the United States without adding increasing amounts of carbon dioxide to the environment. We expect these shortages will take many years to address, and during these years each shortage will exert upward pressure on the overall price level. What is also clear is the Federal Reserve pushing up interest rates to try to temporarily suppress demand will do nothing to address these shortages and would likely lengthen the time needed for the supply adjustments.
How America created low inflation environments
The 1950s and 1960s were an extraordinary period in America. The U.S. dominated global economic activity like it never had in the past or would in the foreseeable future. With only about six percent of the world’s population in 1950, the United States sat at the center of the global trading system and accounted for more than a quarter of the world’s total economic output. The American economy at that point in time was twenty percent larger than the economies of China, Japan, Germany, the U.K. and France combined.
Labor productivity grew at an unusually fast pace as innovation, as measured by total factor productivity, spread rapidly throughout the economy (see table above). Many of the productivity benefits of the intense “learn by doing” processes that occurred during the high-pressure war years, when manufacturing employment increased from 9 million in 1940 to 16.5 million in 1943, were applied to the “civilian” economy after the war. The federal government had invested heavily in refreshing the nation’s capital stock during the war, and private capital spending grew at a solid pace along with strong government spending on infrastructure and housing the 1950s and 1960s.
The cost of housing and utilities increased by an average of just 1.6 percent per year from the mid-1950s to the mid-1960s (see chart). Electricity prices rose by less than one percent a year during the 1950s & 1960s. Coal was the main fuel used for power generation, it was plentiful, and there were limited public concerns about the environmental impacts of burning coal until the late 1960s. Oil prices – the other main source of power in America – increased from $2.57 per barrel at the start of the 1950s to $3.35 per barrel at the end of the 1960s, an average annual increase of 1.3 percent.
Unions were powerful in the post-war decades, a legacy of the Roosevelt Administration’s pro-labor policies as well as the extraordinary “team spirit” that was developed during the war years. That – plus consistently strong demand for labor — were the reasons hourly compensation rose by more than 5 percent per year, on average, significantly raising the living standards of most Americans and keeping GDP advancing at a rapid pace. Strong wage growth fit comfortably within a low inflation society as rapid productivity improvements dampened increases in unit labor costs (table above).
The 2010s were a very different story
The second time when inflation was at or below 2 percent for a cluster of several years was during the decade following the financial crisis. These years looked nothing like the strong growth, high productivity years of the 1950s and 1960s in America. Instead, this decade appears to have been the zenith of the powerful, supply-side deflationary forces that were set in motion by the North American Free Trade Agreement in 1994 and by China’s formal inclusion in the global trading system in 2001.
These two landmark trade agreements catalyzed a 25-year decline in the prices of durable goods in the United States (blue line, chart above) and were associated with a sharp drop in U.S. manufacturing employment (red line, right scale). Durable goods prices declined by a cumulative 38 percent from 1994 to 2019, or an average of 1.3 percent per year – a powerful force pushing down on overall measured price inflation in America.
The “outsourcing” of manufacturing jobs to China and other lower wage countries played a central role in weakening the bargaining power and wage growth of American workers. Average annual increases in hourly compensation declined from 4.0 percent in the 1990s, to 3.7 percent in the 2000s, to just 2.5 percent in the 2010s. The spike in unemployment during the financial crisis, to 10 percent in 2009, was an additional powerful and lingering influence suppressing wage growth.
The boom in residential investment spending in the years leading up to the financial crisis turned into a bust during and after the financial crisis, declining to its lowest share of GDP in the data set (chart above). As the freeze of credit availability slowly faded away, residential investment started growing again, and housing costs – measured by owners’ equivalent rent — moved up to a more normal 3.0-to-3.5 percent annual rate in the second half of the decade (blue line, below). But the price of oil (red line), and overall energy costs, dropped sharply just as annual increases in housing costs were normalizing, so overall measured price inflation stayed very low.
In summary, 2 percent (or less) inflation was achieved in America during the 2010s by: (a) importing durable goods deflation, (b) experiencing anemic growth in wages that held down price increases in the services sector, (c) modest, if any, increases in housing costs, and (d) oil prices that averaged $53 per barrel in the second half of the decade.
Changing behaviors lead to stronger pricing
But each of these trends – and the policies and behaviors that supported them – have changed dramatically in the past few years. Durable goods prices turned up sharply as supply chains snarled just as purchasing power grew rapidly when government transfer payments ballooned in response to the Covid pandemic. More significantly for the future, the political consensus supporting the outsourcing of manufacturing jobs – along with skills and technology – to China has crumbled. Rather than encouraging imports of low-priced goods, now ideas about further import restrictions and/or higher tariffs dominate discussions as policy makers seek to expand domestic production of high value-added goods like electric vehicles and batteries and commodities like steel and aluminum.
This sea change in government policy could lead to sustained increases in the prices of durable goods in coming years, reversing the trend in place since the mid-1990s.
A tight labor market and some tough, high-profile union bargaining have contributed to average growth in hourly compensation in the 2020s of more than 5 percent (see table above). Median wage growth (red line, above) and service price inflation (blue line) have both moderated during the past 12 months, but both measures remain well above the growth rates observed in the 2010s. In a major victory for organized labor, workers at a Volkswagen plant in Tennessee voted overwhelmingly to join the United Automobile Workers union in mid-April, becoming the first nonunion auto plant in a Southern state to do so. While the level of unionization remains close to a forty-year low, it no longer appears to be declining. The stabilization of overall union membership and strong wage growth shouldn’t be surprising, since by several metrics, the Biden Administration is the most pro-labor administration since that of Roosevelt.
Immigration has boomed in recent years to meet the surging demand for labor in the U.S. Foreign born workers (read line, chart below) have filled 100 percent of the net new jobs created since the start of the 2020s. The increase in the working age population associated with this immigration, coupled with a work-from-home inspired “flight to the suburbs,” has turned the excess supply that existed in many housing markets across the U.S. 10-15 years ago, into widespread supply shortages, putting strong upward pressure on house prices and rents. Residential investment spending averaged just 3.9 percent of GDP in 2023, roughly one standard deviation below the 75-year average of 4.6 percent. Bringing supply up in line with demand for housing looks like it could take many years to achieve.
The next five years … and beyond
Going back to the 2010s mix of imported goods deflation, anemic wage growth, and modest annual increases in housing costs now looks like a highly improbable outcome during in the second half of this decade. And a return to a 1950s-type economy seems equally unachievable given weak U.S. demographics, the change in global power dynamics, and roughly 15 years of sluggish productivity growth. So, the likelihood of a “cluster” of low inflation years emerging in the second half of the 2020s looks remote to us. An inflation rate between 3.5 and 5.5 percent may be the best we can hope for during the next five years, like what America experienced in the second half of the 1980s. With big parts of the world all seeking the same minerals and raw materials to make the “energy transition”, risks to inflation do seem skewed to the upside. Big parts of the world at war point in the same direction.
But if we look further out in time, say to the 2030s, an economy with labor productivity growth of 2.0-to-2.5 percent, wage growth of 4.5-to-5.0 percent, capital input growth of about 4.5 percent, and inflation, if we are lucky, of about 4.0 percent, would seem to be a possibility. This scenario would place America somewhere between the two extremes of the 1950s and 2010s.
The Fed’s behavior looks critical to the timing of that type of transition. The Federal Reserve does not have the tools to “fine tune” either the pace of growth or the rate of inflation, but it can definitely tighten financial conditions enough to stop growth and cause employment and investment to weaken dramatically.
It has done it in the past, in the late 1970s and early 1980s. Armed with a fresh mandate from Congress to bring inflation down, the Volcker-led Fed pushed short-term rates (red line, above) up to levels more than 5 percentage points above the pace of nominal GDP growth (red line). But they are nowhere near that point now. Financial conditions remain quite lose in the U.S. (see chart below). Not a real surprise with the Fed funds rate holding steady at a level modestly below the pace of nominal GDP (now 5.5 percent).
If the Fed does tighten financial conditions severely in the pursuit of their illusive 2 percent inflation target, it would delay, but probably not dramatically reduce the likelihood of getting the strong investment, solid productivity growth scenario sketched out above. Given the urgency now being placed on re-industrializing America by both political parties – due to national security concerns, climate concerns, or both — it is doubtful that Congress would tolerate a prolonged period of Fed-induced financial austerity.
How America can successfully re-industrialize
America will need consistently strong growth of capital spending and years of “learning by doing” in both the manufacturing and services sectors to spread innovations widely throughout the economy. And this investment and innovation will need to address the powerful demand for electricity driven by artificial intelligence and data centers more broadly, as well as the transition of our transportation fleets and residential and commercial structures towards more reliance on electricity and less dependence on fossil fuels.
These are herculean tasks, requiring large amounts of capital and large amounts of commodities and raw materials that are now hard to mine and complicated to process. It will also require a lot more labor, primarily through immigration given U.S. demographic patterns. And more labor will require more housing.
The strong investment, solid productivity growth scenario – if it happens — will probably require profound changes in U.S. corporate behavior and elevated government investment levels. One factor contributing to weak growth of capital inputs and sluggish productivity improvements during the past 15 years is a major shift in the allocation of corporate cash flow by CEOs across corporate America. In the 1950s and 1960s, 17 percent of after-tax cash flow was allocated to shareholders, primarily through dividend payments. By the 2010s, this figure was up to 42 percent, as the popularity of share buy-back programs grew in scale. Raising EPS by financial engineering rather than product and process innovation has proven to be very seductive in large parts of corporate America. Growth in capital inputs needs to accelerate.
What about bond yields?
One important aspect of a stronger growth, higher inflation environment than the U.S. lived through in the 2010s is likely to be a wider spread between the 10-year yield and annual consumer price inflation. As inflation climbed higher from the beginning of 1966 to mid-1973, the spread between the 10-year yield (blue line below) and inflation measured by the PCE deflator (red line) averaged 235 basis points. But following the twin inflation peaks in 1974, and then again in 1980, the spread between Inflation and 10-year yields was much wider. Inflation moved up and down between a high of 5.2 percent and a brief low of 1.6 percent between January 1982 to December 1990, and during these years the spread between the 10-year yield and inflation was 600 basis points.
Expectations adjust slowly. We instinctively believe that tomorrow will look a lot like yesterday. But if we are moving into a world where growth is faster and inflation is higher than they were in the 2010s, the spread between spot inflation and bond yields is likely to widen from the current 210 basis points as we move further into the 2020s and beyond.
Douglas Cliggott | Provincetown, MA | 25 April 2024
While we make every effort to ensure that the analysis in this note is as accurate as possible, we do not guarantee that the information contained is either complete or correct. The material has been provided for informational and educational purposes only. The information is not intended to provide or constitute investment, accounting, tax, or legal advice.